mogli

Mogli was created to simplify the way decentral energy installation companies manage their daily operations. Companies installing solar panels, heat pumps, and battery systems often rely on many disconnected software tools to manage customers, proposals, projects, invoices, and operations. Because these systems do not fully communicate with each other, employees spend a large amount of time manually copying the same information between platforms, creating inefficiencies, frustration, and human error.

Over 4 to 5 months, I worked on Mogli alongside the very talented creators and my fellow UX team, focusing on simplifying automation for non technical users in the decentral energy sector.

Mogli’s challenge was clear: energy installation companies relied on multiple disconnected tools to manage customers, projects, invoices, and operations, forcing teams to repeatedly enter the same information across platforms. While automation could solve this, existing tools often felt too technical and overwhelming for everyday users.

We started with user research and interviews to understand how people experienced current automation platforms. One insight appeared consistently: users struggled with drag and drop workflow systems and often did not trust whether their automations would work correctly. Many ended up manually double checking processes, which defeated the purpose of automation.

This helped us realize the problem was not only usability, but also trust. Users wanted automation to feel simple, supportive, and easy to understand. Our problem statement: how can we help busy, non technical users create automations in a way that feels simple, clear, and trustworthy?

To organize these insights, we created personas, user journey maps, and workflow structures based on real user behaviors. During interviews, we noticed users naturally described tasks in plain language rather than technical logic, saying things like, “When an order is completed, create the invoice and send it to the customer.”

That observation led us to the final direction of the product: instead of another drag and drop builder, we designed an AI chatbot experience where users could create automations through natural conversation. We also introduced confirmation steps and transparent feedback so users could clearly understand what the AI was doing and feel more confident using it.

From there, we developed low fidelity wireframes, mid fidelity concepts, and a high fidelity clickable prototype, testing and refining the experience through user feedback.

The final result was an AI powered chatbot interface that made workflow automation feel more natural, intuitive, and accessible for non technical users.